AI Paper Atlas

Research map / Architectures and efficiency / RNNs and normalization

House of Cans: Covert Transmission of Internal Datasets via Capacity-Aware Neuron Steganography

Xudong Pan, Shengyao Zhang, Mi Zhang, Yifan Yan, Min Yang

NeurIPS 2022 · Architectures and efficiency › RNNs and normalization · approach: Efficient network compression and training · 1 citations (OpenAlex)

Read the paperCheck your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
increase stolen samples from 512 to 4096Cans · SpeechCommandMSE: 2.7 × 10−4 -> 4.8 × 10−4Hurt
increase stolen samples from 512 to 4096Cans · SpeechCommandcarrier Delta Perf: −0.43% -> −0.35%No clear effect
steal three datasets simultaneously instead of Speech aloneCans · SpeechCommand 1024 samplesMSE: 3.5 × 10−4 -> 4.1 × 10−4No clear effect
increase FaceScrub stolen images from 1024 to 32768Cans · FaceScrubSSIM: 1.0 -> 0.6Hurt
weight pruning ratio beta 0.3 on carrier modelCans · FaceScrub 512 imagesSSIM of decoded images: 0.994 -> 0.8Hurt
weight pruning ratio beta 0.3 on carrier modelCans · FaceScrub 512 imagescarrier performance: over 10%Hurt
remove fusion (weight pool restoration) from pruned model decoding (ablation)Cans · FaceScrub pruningsecret model reconstructionHurt
use Cans instead of Sign/Covariance/LSB baselines at N=64+Cans · FaceScrubhiding feasibility: Nmax = 76Improved
use LSB encoding at N=64 instead of CansLSB baseline · FaceScrubcarrier accuracyHurt
use Cans instead of Sign and Covariance encodingCans · FaceScrub/CIFAR-10SSIMImproved

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy